In computer or media systems having a source unit coupled to a display device by a cable, it has become desirable to provide audio output at the location of the display device. When this is achieved, the audio output may be presented to the user via one or more sound transducers that are typically integrated into the display device or otherwise placed in close proximity to the display device.
One known technique for providing audio output at the location of the display device is to use two separate cables between the source unit and display device-one cable for video signals and one cable for audio signals. For example, it is known to use one VGA (“Video Graphics Array”) cable and pair of connectors to couple analog video signals and associated video timing signals from the source unit to the display device, and to use a separate non-VGA cable with its own pair or pairs of connectors to couple analog audio signals from the source unit to the display device. It is also known to use a DVI (“Digital Visual Interface”) cable and pair of connectors to couple analog or digital video signals and associated video timing signals from the source unit to the display device, and to use a separate non-DVI cable with its own pair or pairs of connectors to couple analog audio signals from the source unit to the display device. Typically, two channels of audio are provided using the separate-cables technique, the left channel and the right channel. Although the separate-cables technique represents a straightforward solution, it is expensive because it requires two cables and at least two pairs of connectors, and it degrades the quality of the audio somewhat because the audio signals are transmitted in analog form.
Another known technique for providing audio output at the location of the display device is to follow the HDMI (“High-Definition Multimedia Interface”) standard. The HDMI standard specifies a way of coupling digital audio and digital video signals between the source unit and display device in time-multiplexed fashion over a single cable and pair of connectors. An HDMI cable includes enough wires to support three different types of communication channels: TMDS (“Transition Minimized Differential Sampling”), DDC (“Display Data Channel”) and CEC (“Consumer Electronics Control”). More specifically, an HDMI cable includes four shielded TMDS wire pairs. These four TMDS wire pairs are used to transmit digital video and timing signals from the source unit to the display device. These same four TMDS wire pairs are also used to transmit digital audio signals from the source unit to the display device during the horizontal and vertical blanking intervals of the video. To accomplish this, the HDMI standard specifies a new transport protocol that is superimposed on top of the older (“International Electrotechnical Commission”) IEC 60958 and IEC 61937 digital audio transmission protocols (hereinafter the “IEC protocols”).
According to the HDMI transport protocol, audio data being carried across the TMDS links does not retain the original audio sample clock that the IEC protocols provided. Instead, typically between one and four packets of audio data are transmitted over the TMDS links during every horizontal blanking interval, each of the packets carrying a payload of up to four subpackets, each of the subpackets carrying at most one IEC protocol frame or block of audio samples. The sound system associated with the display device must then continually recreate the audio sample clock and synchronize it with the video clock using HDMI-specified audio clock regeneration packets that are also transmitted across the TMDS links.
Because numerous audio encodings, sample rates and sample sizes are supported by the IEC protocols, it is necessary to set corresponding parameters for use between a given source unit and display device/audio system. According to the HDMI standard, the source unit uses the DDC channel in the HDMI cable to determine which audio encodings are supported by the display device's audio system. It does so by reading an HDMI-specified “vendor specific data block” from the E-EDID (“Enhanced Extended Display Identification Data”) data stored in the display device. The source unit then dictates which parameters will be used for audio by setting appropriate bits in the IEC protocol packets and by sending HDMI-specified “audio infoframe packets” at frequent, HDMI-specified times.
One of the benefits of the HDMI standard is that it addresses the problem of audio quality degradation by transmitting digital audio signals instead of analog audio signals. The HDMI standard also eliminates the need for separate audio and video cables and pairs of connectors by time-multiplexing digital audio and digital video on a common set of TMDS wires. Unfortunately, HDMI implementations introduce significant manufacturing expense: The relatively elaborate HDMI cable and connectors are costly, and the encoding/decoding logic that is necessary to implement the HDMI protocol in the source unit and in the display device are complex. Moreover, because HDMI sends audio packets during video blanking intervals, the bandwidth available for audio in an HDMI implementation is limited by, and various in accordance with, the video timing of the display device and the source unit's graphics system. In addition, many display devices and source units that support the HDMI standard also support the VGA and/or the DVI standards for backward compatibility. Consequently, the expense of including VGA and DVI connectors is not avoided in such HDMI-equipped systems.